1 /*- 2 * Copyright (c) 1992, 1993, 1994 Henry Spencer. 3 * Copyright (c) 1992, 1993, 1994 4 * The Regents of the University of California. All rights reserved. 5 * 6 * This code is derived from software contributed to Berkeley by 7 * Henry Spencer. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 3. Neither the name of the University nor the names of its contributors 18 * may be used to endorse or promote products derived from this software 19 * without specific prior written permission. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 24 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 31 * SUCH DAMAGE. 32 * 33 * @(#)regexec.c 8.3 (Berkeley) 3/20/94 34 */ 35 36 #if defined(LIBC_SCCS) && !defined(lint) 37 #if 0 38 static char sccsid[] = "@(#)regexec.c 8.3 (Berkeley) 3/20/94"; 39 #else 40 static char rcsid[] = "$OpenBSD: regexec.c,v 1.8 2003/06/02 20:18:36 millert Exp $"; 41 #endif 42 #endif /* LIBC_SCCS and not lint */ 43 44 /* 45 * the outer shell of regexec() 46 * 47 * This file includes engine.c *twice*, after muchos fiddling with the 48 * macros that code uses. This lets the same code operate on two different 49 * representations for state sets. 50 */ 51 #include <sys/types.h> 52 #include <stdio.h> 53 #include <stdlib.h> 54 #include <string.h> 55 #include <limits.h> 56 #include <ctype.h> 57 #include <regex.h> 58 59 #include "utils.h" 60 #include "regex2.h" 61 62 /* macros for manipulating states, small version */ 63 #define states long 64 #define states1 states /* for later use in regexec() decision */ 65 #define CLEAR(v) ((v) = 0) 66 #define SET0(v, n) ((v) &= ~((unsigned long)1 << (n))) 67 #define SET1(v, n) ((v) |= (unsigned long)1 << (n)) 68 #define ISSET(v, n) (((v) & ((unsigned long)1 << (n))) != 0) 69 #define ASSIGN(d, s) ((d) = (s)) 70 #define EQ(a, b) ((a) == (b)) 71 #define STATEVARS long dummy /* dummy version */ 72 #define STATESETUP(m, n) /* nothing */ 73 #define STATETEARDOWN(m) /* nothing */ 74 #define SETUP(v) ((v) = 0) 75 #define onestate long 76 #define INIT(o, n) ((o) = (unsigned long)1 << (n)) 77 #define INC(o) ((o) <<= 1) 78 #define ISSTATEIN(v, o) (((v) & (o)) != 0) 79 /* some abbreviations; note that some of these know variable names! */ 80 /* do "if I'm here, I can also be there" etc without branches */ 81 #define FWD(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) << (n)) 82 #define BACK(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) >> (n)) 83 #define ISSETBACK(v, n) (((v) & ((unsigned long)here >> (n))) != 0) 84 /* function names */ 85 #define SNAMES /* engine.c looks after details */ 86 87 #include "engine.c" 88 89 /* now undo things */ 90 #undef states 91 #undef CLEAR 92 #undef SET0 93 #undef SET1 94 #undef ISSET 95 #undef ASSIGN 96 #undef EQ 97 #undef STATEVARS 98 #undef STATESETUP 99 #undef STATETEARDOWN 100 #undef SETUP 101 #undef onestate 102 #undef INIT 103 #undef INC 104 #undef ISSTATEIN 105 #undef FWD 106 #undef BACK 107 #undef ISSETBACK 108 #undef SNAMES 109 110 /* macros for manipulating states, large version */ 111 #define states char * 112 #define CLEAR(v) memset(v, 0, m->g->nstates) 113 #define SET0(v, n) ((v)[n] = 0) 114 #define SET1(v, n) ((v)[n] = 1) 115 #define ISSET(v, n) ((v)[n]) 116 #define ASSIGN(d, s) memcpy(d, s, m->g->nstates) 117 #define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0) 118 #define STATEVARS long vn; char *space 119 #define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \ 120 if ((m)->space == NULL) return(REG_ESPACE); \ 121 (m)->vn = 0; } 122 #define STATETEARDOWN(m) { free((m)->space); } 123 #define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates]) 124 #define onestate long 125 #define INIT(o, n) ((o) = (n)) 126 #define INC(o) ((o)++) 127 #define ISSTATEIN(v, o) ((v)[o]) 128 /* some abbreviations; note that some of these know variable names! */ 129 /* do "if I'm here, I can also be there" etc without branches */ 130 #define FWD(dst, src, n) ((dst)[here+(n)] |= (src)[here]) 131 #define BACK(dst, src, n) ((dst)[here-(n)] |= (src)[here]) 132 #define ISSETBACK(v, n) ((v)[here - (n)]) 133 /* function names */ 134 #define LNAMES /* flag */ 135 136 #include "engine.c" 137 138 /* 139 - regexec - interface for matching 140 = extern int regexec(const regex_t *, const char *, size_t, \ 141 = regmatch_t [], int); 142 = #define REG_NOTBOL 00001 143 = #define REG_NOTEOL 00002 144 = #define REG_STARTEND 00004 145 = #define REG_TRACE 00400 // tracing of execution 146 = #define REG_LARGE 01000 // force large representation 147 = #define REG_BACKR 02000 // force use of backref code 148 * 149 * We put this here so we can exploit knowledge of the state representation 150 * when choosing which matcher to call. Also, by this point the matchers 151 * have been prototyped. 152 */ 153 int /* 0 success, REG_NOMATCH failure */ 154 regexec(preg, string, nmatch, pmatch, eflags) 155 const regex_t *preg; 156 const char *string; 157 size_t nmatch; 158 regmatch_t pmatch[]; 159 int eflags; 160 { 161 register struct re_guts *g = preg->re_g; 162 #ifdef REDEBUG 163 # define GOODFLAGS(f) (f) 164 #else 165 # define GOODFLAGS(f) ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND)) 166 #endif 167 168 if (preg->re_magic != MAGIC1 || g->magic != MAGIC2) 169 return(REG_BADPAT); 170 assert(!(g->iflags&BAD)); 171 if (g->iflags&BAD) /* backstop for no-debug case */ 172 return(REG_BADPAT); 173 eflags = GOODFLAGS(eflags); 174 175 if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags®_LARGE)) 176 return(smatcher(g, (char *)string, nmatch, pmatch, eflags)); 177 else 178 return(lmatcher(g, (char *)string, nmatch, pmatch, eflags)); 179 } 180